The broad long term objectives are twofold. One is to develop methods that will extend the power of solution NMR beyond the study of native globular proteins to the study of nonnative and nonglobular proteins. The second objective is to use NMR to understand protein folding mechanisms by studying structures of partially folded proteins, and by assessing the effect of amino acid substitutions on structure, stability and folding kinetics. These studies will form the basis for using NMR to understand how interruptions in the Gly-X-Y pattern, found in collagen diseases like Osteogenesis Imperfecta and Ehlers Danlos Syndrome, can result in serious disease. The first aim is to characterize the partially folded state of guinea pig alpha-lactalbumin to elucidate the nature of protein folding intermediates. More specifically, we wish to learn which regions of the molten globule state contain regions of secondary structure and whether tertiary interactions are important in stabilizing these regions of secondary structure. We will characterize the partially folded state by 1H NMR methods, and by isotope labelling and heteronuclear 2D and 3D NMR experiments. Key mutants will be made to assess the effects of sequence change on secondary structure and tertiary interactions of the partially folded state. The second aim is to obtain, for the first time, individual residue assignments and the NMR solution structure of triple helical peptides, to determine the role of individual amino acids in stabilizing the triple helix and to understand how key residues direct protein folding. We will examine the effects, by 1D NMR, of (Gly-X-Y) sequence changes on the amount of triple helix formed, and on the kinetics and thermodynamics of folding. To obtain the solution structure we will first design synthetic triple helical peptides to facilitate the spin system identification process. Then we propose 1H NMR experiments as well as heteronuclear NMR experiments that should allow us to perform sequential resonance assignments, and distinguish inter from intra strand NOE's.